The primary objective is to evaluate the possibility that non-pharmacologic manipulation of the autonomic nervous system in normal individuals later modifies the risk for sudden cardiac death. This is a logical consequence of the concept that high cardiac sympathetic activity is arrhythmogenic and favors the onset of ventricular fibrillation (VF); high cardiac vagal activity during acute myocardial ischemia is antifibrillatory and may be life-saving. The study will be performed using an established chronic animal model for sudden cardiac death developed by the investigators. Conscious dogs perform a sub-maximal exercise stress test; at the beginning of the last minute of exercise, when heart rate os 210-220 b/min, a two minute occlusion of the circumflex coronary artery is performed. After on e minute of occlusion, exercise stops an occlusion is maintained for an additional minute. With this protocol, VF occurs in 40% of the animals being largely dependent on autonomic reflex responses. In the present study, exercise training will be used to induce specific changes in the autonomic reflex control of the heart, i.e. to augment vagal tone reflexes. A depressed baroreflex sensitivity (BRS), largely a marker of weak vagal reflexes, is associated with an increased risk for sudden death even in normal individuals. Accordingly, an initial risk stratification will be performed based on three different levels of BRS (one, less that 15; two, between 15 and 20; and three, greater than 20 msec/mmHg) and the animals randomized into two treatment groups. One group will undergo exercise training for six weeks while the other serves as a cage rested control for an equivalent period of time. After six weeks, risk for sudden death will be added based on the outcome of the evident in dogs with depressed BRS. The effect of exercise training on the electrophysiologic characteristics of the heart and on its vulnerability to VF will be tested by determining ventricular refractoriness and the threshold for repetitive extrasystole. Autonomic blockade, by atropine and/or propranolol, or stimulation by the muscarinic against oxotremorine, will be used to partition the autonomic components largely involved in the specific outcome: VF vs survival. There would be obvious and major health related implications should exercise training in healthy individuals reduce the later risk for sudden cardiac death during an ischemia event.